
Manual – Starflow Ultrasonic Doppler Instrument
Model 6526
Unidata Manual - 6526 Starflow User Manual Issue 17 01 2020.docx
Page 50
These result in frequency shifts being measured that are outside the range of velocities expected.
They can represent several percent of measurements in a scan but are typically widely distributed.
They are usually statistically irrelevant because of the signal analysis techniques used to define the
median.
12.2
Channel Characteristics Affecting the Signal
All signals are affected by the factors discussed above. Additional factors related to channel and
flow characteristics can influence the measured velocity. These result in a widening of the
histogram base because of the broader spread of Doppler shifts measured.
In a channel with laminar flows the shape of the peak remains symmetrical. If other factors disturb
the velocity profiles within the field of view of the Starflow, the histogram may become distorted.
This may introduce a bias to the computed velocity.
If the shape of the histogram is the same from cycle to cycle then it represents the Starflow view of
the channel and not some transient aberration. Hydrographers have to decide if this represents a
valid velocity measurement, or if the site is unsuitable.
12.2.1
The Nature and Location of the “Reflectors” in the Water
The Starflow measures echoed Doppler shifts from anything that is acoustically reflective and in
the signal beam. It cannot discriminate where the echo is coming from and the bigger, closer
reflectors will generate a stronger signal than smaller or more distant reflectors.
Reflectors can include:
•
All types of particles in suspension.
•
Alluvial saltation load such as sand, pebbles and rocks moving at or near the bed.
•
Leaves, sticks, and organic material.
•
Surface waves from wind or velocity.
•
Bubbles and entrapped air pockets.
During the time it takes to traverse the signal beam a particularly good reflector can be measured
100 or more times. The velocity of this reflector may bias or dominate one velocity measurement.
This will tend to be averaged out during the logging period if it only happens occasionally.
If it happens consistently it is a site characteristic. The relevance of this to the mean channel
velocity will need to be understood and compensated for in processing.